STATEWIDE TARGET FISH COMMUNITY ASSESSMENT PEMIGEWASSET - FINAL REPORT

Prepared for: Prepared by:

July, 2018

Table of Contents

I. Introduction ...... 1

II. Designated River Delineation ...... 1 Delineation Methods ...... 1 Delineation Results ...... 7

III. Reference River Data Selection ...... 13 Reference River Selection Methods ...... 13 Reference River Selection Results ...... 17

IV. TFC Model Development ...... 33 TFC Model Development Methods ...... 33 TFC Model Results ...... 34

V. References Cited ...... 42

Appendix A ...... 43

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Report i July, 2018

List of Tables

Table - 1: Fish species removed from the NMDS delineation analysis...... 3 Table - 2: GIS layers used in the delineation analysis...... 4 Table - 3: Gradient and watershed size class categories from the Northeast Aquatic Habitat Classification System layer...... 4 Table - 4: Soil types used in the delineation analysis...... 5 Table - 5: Delineation parameter descriptions and break justifications...... 7 Table - 6: Information pertaining to reaches delineated by the TFC breaks...... 8 Table - 7: Elevation classes and descriptions from the Northeast Aquatic Habitat Classification System layer...... 13 Table - 8: Chemical classes and descriptions from the Northeast Aquatic Habitat Classification System layer...... 13 Table - 9: Temperature classes and descriptions from the Northeast Aquatic Habitat Classification System layer...... 14 Table - 10: Characteristics used to select reference for the upper segment...... 17 Table - 11: Species counts for reference rivers for the upper segment...... 21 Table - 12: Characteristics used to select reference rivers for the middle segment...... 22 Table - 13: Species counts for reference rivers for the middle segment...... 26 Table - 14: Characteristics used to select reference rivers for the lower segment...... 28 Table - 15: Species counts for reference rivers for the lower segment...... 32 Table - 16: Comprehensive list of native species used for the Designated River watershed, as determined from the greater basin area...... 34 Table - 17: Count of fish from reference river data and expected percentage (TFC Model) of species for the upper segment...... 37 Table - 18: Count of fish from reference river data and expected percentage (TFC Model) of species for the middle segment...... 39 Table - 19: Count of fish from reference river data and expected percentage (TFC Model) of species for the lower segment...... 41 Table – A1: List of common and scientific names for fish species in the fish community sample dataset (includes samples from NY, CT, RI, MA, VT, NH, and ME)...... 44

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report ii July, 2018

List of Figures

Figure - 1: Gradient (left panel) and order (right panel), along with fish sampling locations...... 9 Figure - 2: Soils (left panel), watershed size class (right panel), and NH predicted fish community types (right panel)...... 10 Figure - 3: Bedrock composition and water chemical classification...... 11 Figure - 4: Delineated segments derived from the TFC break points...... 12 Figure - 5: Theoretical example of evaluating the number of samples (i.e. whether data are sufficient for further analysis) using MultSE...... 16 Figure - 6: Initial selection of reference river fish community samples for the upper segment...... 18 Figure - 7: MultSE (beginning at n=2) of fish community data for reference rivers initially selected for the upper segment...... 19 Figure - 8: MultSE for the final reference river selection for the upper segment...... 20 Figure - 9: Initial selection of reference river fish community samples for the middle segment...... 23 Figure - 10: MultSE (beginning at n=2) of fish community data for reference rivers initially selected for the middle segment...... 24 Figure - 11: MultSE for the final reference river selection for the middle segment...... 25 Figure - 12: Initial selection of reference river fish community samples for the lower segment...... 29 Figure - 13: MultSE (beginning at n=2) of fish community data for reference rivers initially selected for the lower segment...... 30 Figure - 14: MultSE for the final reference river selection for the lower segment...... 31 Figure - 16: Graphical representation of the TFC Model for the upper segment...... 36 Figure - 17: Graphical representation of the TFC Model for the middle segment...... 38 Figure - 18: Graphical representation of the TFC Model for the lower segment...... 40

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report iii July, 2018

I. Introduction

Target Fish Community models have been developed for a number of instream flow related projects in the Northeast. The TFC development process, as defined by Bain and Meixler (2005), uses fish community data from the best available reference rivers that would characterize a feasible and currently relevant fish community. As such, the TFC model does not represent a historically “natural” community, but instead represents a community that would be expected to exist in the present time given relatively low direct anthropogenic impact on instream habitat. This approach has been useful for evaluating the biological integrity of and rivers by comparing the existing fish community with that of the predicted TFC. NHDES plans to use the TFC models in support of development of protected instream flows on Designated Rivers, and also as guidelines for evaluating the biological integrity.

Across the State, there is the potential for wide variability in hydromorphologic and geologic features that would have shaped the natural fish community. Additionally, sections of the Designated Rivers may be geomorphologically different from other sections, and may have naturally supported different fish communities. Therefore, the Designated Rivers must first be delineated prior to TFC development. The goal of the delineation was to segment Designated Rivers with the fewest possible breaks based on fish community shifts on a watershed scale that are relevant to NHDES management goals. After delineation, suitable fish community data from reference rivers that are geomorphologically similar to each delineated segment were selected for potential use in the TFC model using an iterative GIS and data screening process. Once reference river data were thoroughly screened, TFC models were developed using the Bain and Meixler (2005) methodology. II. Designated River Delineation

DELINEATION METHODS Delineation of the Designated Rivers into segments was based on a combination of datasets, as described in more detail below, including current fish community data, predicted fish community types, and a variety of GIS layers that would allow for visualization of changes in stream geomorphology and overall character. The exact location of river segment delineation was based on a combination of factors that would lead to shifts in fish communities that may pertain to management of instream flow and habitat.

FISH COMMUNITY DATA ASSESSMENT Fisheries sampling data, as provided from NHDES for the Designated River watersheds (including many sites that were not directly on Designated Rivers), were determined to be suitable for further comparative analysis if they were collected by electrofishing in non-impounded, riverine reaches. These sites were used to develop a site-species matrix for Nonmetric Multidimensional Scaling (NMDS) ordination. NMDS ordination provides a visual display of sites, with the locations of the sites in ordination space based on the community of species present; this is useful for determining similarities and differences of fish communities among sites.

Species captured at fewer than five locations within the entire dataset were removed from this matrix (Table - 1). This was found to be necessary for the NMDS ordination to reach a solution. In general, first- pass count data were used for further analysis for sites with greater than 50 individual fish captured, and more than six species present. However, all sites from within the Designated River segments were kept

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 1 July, 2018

in the dataset, even if they did not meet the individual or species criteria. These data were used with caution and were only used to visualize differences in the ordination results. The final matrix included 157 site locations and 35 species.

The matrix was converted to catch per unit of effort (CPUE) at each location prior to NMDS development using ratio estimation techniques from Hansen et al. (2007), which is represented as:

∑푛 푦 ̂ 푖=1 푖 퐶푃푈퐸 = 푅 = 푛 ∑푖=1 푥푖

Where the CPUE ratio estimator (푅̂) equals the sum of the catch (푦푖) at all sites sampled, divided by the sum of the effort (seconds of electrofishing) at all sites sampled (푥푖).

The NMDS ordination was developed using the Statistical Package R and the metaMDS function. The metaMDS function was set to automatically transform the data matrix, which optimizes data standardization based on the data structure. The ordination was based on Bray-Curtis dissimilarity, which calculates how different community data from various sites are based on proportional abundance, and is considered to be the most reliable distance measure for NMDS ordination of community structure (Clarke 1993).

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 2 July, 2018

Table - 1: Fish species removed from the NMDS delineation analysis.

Scientific Name Common Name Locations in Dataset

Chrosomus neogaeus Finescale Dace 1 Clinostomus funduloides Rosyside Dace 1 Coregonus clupeaformis Lake Whitefish 1 Fundulus heteroclitus Mummichog 1 Notropis heterolepis Blacknose Shiner 1 Pimephales notatus Bluntnose Minnow 1 Pimephales promelas Fathead Minnow 1 Pomoxis annularis White Crappie 1 Prosopium cylindraceum Round Whitefish 1 Alosa aestivalis Blueback Herring 2 Alosa pseudoharengus Alewife 2 Alosa sapidissima American Shad 2 Fundulus diaphanus Banded Killifish 2 Morone americana White Perch 2 Notropis volucellus Mimic Shiner 2 Esox lucius Northern Pike 3 Petromyzon marinus Sea Lamprey 3 Sander vitreus Walleye 3 Pomoxis nigromaculatus Black Crappie 4 Note: Only includes data from NH Designated River Watersheds

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 3 July, 2018

GIS ANALYSIS In addition to the fish community analysis, several physical and biological factors were evaluated to determine if river delineation was appropriate. Break locations were identified at areas where significant changes in stream character, based on available data, occur. Datasets utilized for these analyses are shown in Table - 2. In general, the greatest weight was given to data that are more easily quantifiable, such as Stream Order, Gradient, and Watershed Size, with the other types of data used secondarily to provide any additional support for break locations.

Table - 2: GIS layers used in the delineation analysis. GIS Layer Source Uses Designated Rivers NHDES Stream Order Gradient Northeast Aquatic Habitat Classification TNC Size Class (Watershed Size) Water Chemistry Level III Ecoregion USEPA Ecoregion Predicted Fish Community Type NH Fish Community Types NHDES (cold, warm, transitional) Soils GRANIT General Soil/Geology Lithology USGS/GRANIT General Bedrock Geology Visual Assessment of Character; Orthoimagery ESRI1 Map Background

The Designated River layer was used to determine the extent of the Designated River, and to determine the locations at which the rivers increased in stream order. Changes in gradient and watershed size class were evaluated using the Northeast Aquatic Habitat Classification layer (Table - 3). This layer was developed by The Nature Conservancy (TNC) via the Northeast Aquatic Habitat Classification System Project (Olivero and Anderson 2008).

Table - 3: Gradient and watershed size class categories from the Northeast Aquatic Habitat Classification System layer.

Gradient Class Gradient (%) Size Class Watershed Size (mi2) Very Low < 0.02 Headwater 0 < 3.861 Low >= 0.02 < 0.1 Creek >= 3.861 < 38.61 Low-Moderate >= 0.1 < 0.5 Small River >= 38.61 < 200 Moderate-High >= 0.5 < 2 Medium Tributary River >= 200 < 1,000 High >= 2 < 5 Medium Mainstem River >= 1,000 < 3,861 Very High > 5 Large River >= 3,861 < 9,653

1 The source for the imagery used is cited as: Esri, DigitalGlobe, GeoEye, Earthstar Geographics, CNES/Airbus DS, USDA, USGS, AEX, Getmapping, Aerogrid, IGN, IGP, swisstopo, and the GIS User Community. The source for topographic background is cited as: Esri, HERE, DeLorme, Intermap, increment P Corp., GEBCO, USGS, FAO, NPS, NRCAN, GeoBase, IGN, Kadaster NL, Ordnance Survey, Esri Japan, METI, Esri China (Hong Kong), swisstopo, MapmyIndia, © OpenStreetMap contributors, and the GIS User Community.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 4 July, 2018

Though the NEAHC dataset also includes a temperature classification, for delineation, temperature criteria were evaluated based on the modeled NH Fish Community types, which predicts general fish community types based on , latitude, elevation, and upstream drainage area. As such, TFC Breaks based on predicted coldwater, transitional, and warmwater communities were examined.

Changes in soils and bedrock geology can result in changes in general substrate types and water quality conditions along the stream gradient. The soils layer was examined spatially, using the generic groups provided in the data layer as described in Table - 4.

Table - 4: Soil types used in the delineation analysis.

Group Definition Deeper, loamy textured, moderately well and well-drained IA soils Sandy or loamy over sandy textures; slightly less fertile than IB Group IA Outwash sands and gravels, somewhat excessively to IC excessively drained and moderately well drained Similar to Groups IA and IB, except physically limited due to IIA steep slopes, bedrock outcrops, etc. Poorly drained soils, seasonal high water table, generally IIB within 12 feet of surface

Similarly, patterns in bedrock geology were examined. Underlying bedrock can affect valley and stream shape. Bedrock type also affects the buffering capacity and pH of the stream, which can impact fish communities on a watershed scale. Weathering of different types of bedrock yields different shapes and sizes of material, which in turn could affect substrate and sediment types in the stream, affecting fish on a microhabitat scale. Because the effects on stream fish communities can occur over multiple spatial scales, this analysis focuses on whether general changes in bedrock geology occur along the stream length rather than to characterize specific changes that may occur within a fish community based on bedrock type. Additionally, the water chemistry layer classification of the NEAHC dataset was overlain over bedrock types. Classifications provided by the layer include: 1) Highly Buffered, Calcareous; 2) Moderately Buffered, Neutral; 3) Low Buffered, Acidic; and 4) Assumed Moderately Buffered for Size 3+ Rivers.

Ecoregions have been developed by the U.S. Environmental Protection Agency (EPA) to denote areas of similarity in biotic, abiotic, terrestrial and aquatic ecosystem components, with humans considered as part of the biota. Abiotic and biotic factors included in ecoregion development were geology, physiography, vegetation, climate, soils, land use, wildlife, and hydrology. Only two Level III ecoregions are present in New Hampshire; the Northeastern Highlands (Ecoregion 58) and the Northeastern Coastal Zone (Ecoregion 59). Level III Ecoregions were used in the analysis as a broad descriptor of where a variety of changes may occur in the landscape, and therefore potentially along the stream length.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 5 July, 2018

The Northeastern Highlands, according to the USEPA (2013) description is:

“The Northeastern Highlands cover most of the northern and mountainous parts of New England as well as the Adirondacks and higher Catskills in New York. It is a relatively sparsely populated region characterized by hills and mountains, a mostly forested land cover, nutrient-poor soils, and numerous high-gradient streams and glacial lakes. Forest vegetation is somewhat transitional between the boreal regions to the north in Canada and the broadleaf deciduous forests to the south. Typical forest types include northern hardwoods (maple-beech-birch), northern hardwoods/spruce, and northeastern spruce-fir forests. Recreation, tourism, and forestry are primary land uses. Farm-to-forest conversion began in the 19th century and continues today. In spite of this trend, alluvial valleys, glacial lake basins, and areas of limestone-derived soils are still farmed for dairy products, forage crops, apples, and potatoes. Many of the lakes and streams in this region have been acidified by sulfur depositions originating in industrialized areas upwind from the ecoregion to the west.”

The Northeastern Coastal Zone, according to the USEPA (2013) description is:

“Similar to the Northeastern Highlands (58), the Northeastern Coastal Zone contains relatively nutrient poor soils and concentrations of continental glacial lakes, some of which are sensitive to acidification; however, this ecoregion contains considerably less surface irregularity and much greater concentrations of human population. Landforms in the region include irregular plains, and plains with high hills. Appalachian oak forests and northeastern oak-pine forests are the natural vegetation types. Although attempts were made to farm much of the Northeastern Coastal Zone after the region was settled by Europeans, land use now mainly consists of forests, woodlands, and urban and suburban development, with only some minor areas of pasture and cropland.”

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 6 July, 2018

DELINEATION RESULTS The Pemigewasset River flows approximately 64 miles from its headwaters at , in State Park, to its confluence with the in Franklin, NH, which forms the . The entire river, with the exception of a 10-mile segment in Lincoln and Woodstock, NH, is Designated.

Table - 5: Delineation parameter descriptions and break justifications. Parameter Description and/or Break Justifications

Fish Community (NMDS) No fisheries survey data for analysis.

Stream Order 1st to 6th Order stream, with long segments of 5th and 6th Order stream.

Gradient The upper portion of the river is characterized by high gradients. Much of the 5th Order portion has low/moderate gradients. The 6th Order portion has primarily very low gradient.

Watershed Size Most of the river length is classified as a Medium Tributary River, becoming a Medium Mainstem River near the lower end and containing headwater/creek areas in the upper portions.

NH Fish Community Types Coldwater to transitional in the upper primary segment. Warmwater in the lower primary segment, but surrounded by coldwater regions in the 5th Order portion, becoming surrounded by transitional in the 6th Order portion.

Soils/Bedrock Group IB soils most prevalent in upper primary segment, and in the Geology/Water Chemistry downstream areas of the lower primary segment. Upper areas of the lower primary segment contain more Group IIA soil areas. Granite is the dominant bedrock type in the most upstream parts of the river. Otherwise, the river flows through a variety of bedrock types. Much of the river length was assumed to be moderately buffered given its size, though the upper portions were classified as low buffered/acidic.

Level III Ecoregion No Ecoregion break present.

Other Orthoimagery suggests that the upper portion of the lower primary segment is shallow and relatively sinuous, with many shallow bars on the inside of the bends in the river. This pattern continues downstream nearly to the confluence with the (~2.3 miles below the confluence with the Baker River where the stream becomes 6th Order), below which the river appears to widen and deepen.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 7 July, 2018

TFC Break (Upper): The confluence with the East Branch of the Pemigewasset River, where the river becomes a 5th Order stream. A break at this location would be supported by stream order, watershed size classification, gradient, predicted fish community type, and possibly geology.

TFC Break (Lower): The confluence with the Baker River, where the river becomes a 6th Order stream. A break at this location would be supported not only by stream order, but also by gradient and predicted fish community types in the surrounding landscape.

Table - 6: Information pertaining to reaches delineated by the TFC breaks. TFC Reach Length (miles) Description

Upper 10.15 The headwater areas to the confluence with the East Branch of the Pemigewasset River.

Middle 22.8 The East Branch of the Pemigewasset River to the confluence with the Baker River.

Lower 31.5 From the confluence with the Baker River to the lower end of the river (confluence with the Winnipesaukee River). Consider merging with the Upper Merrimack TFC Reach, which appears to have similar characteristics (see Merrimack River section).

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 8 July, 2018

Figure - 1: Gradient (left panel) and stream order (right panel), along with fish sampling locations.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 9 July, 2018

Figure - 2: Soils (left panel), watershed size class (right panel), and NH predicted fish community types (right panel).

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 10 July, 2018

Figure - 3: Bedrock composition and water chemical classification. .

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 11 July, 2018

Figure - 4: Delineated segments derived from the TFC break points. Upper reach = 10.15 miles; Middle reach = 22.8 miles; Lower reach = 31.5 miles

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 12 July, 2018

III. Reference River Data Selection

REFERENCE RIVER SELECTION METHODS Reference river data were selected initially by using GIS tools, followed by a statistical screening evaluation. GIS ANALYSIS – REFERENCE RIVER AND FISH COMMUNITY SAMPLE SELECTION Segments of reference rivers were selected from the Northeast Aquatic Habitat Classification layer. The selection was based on a set of the following attributes2 defined for each Designated River segment:

 Watershed Size Class (Table 3)  Gradient Class (Table 3)  Elevation Class (Table 7)  Chemical Class (Table 8)  Temperature Class (Table 9)  Level III Ecoregion

Table - 7: Elevation classes and descriptions from the Northeast Aquatic Habitat Classification System layer. Elevation Class Description Elevation (ft) 1 Coastal Zone < 20 2 Low Elevation 20 - 800 3 Mid-to-Lower Elevation Transitional 800 - 1,700 4 Mid-to-Upper Elevation Transitional 1,700 - 2,500 5 High Elevation 2,500 - 3,600 6 Subalpine/Alpine > 3,600

Table - 8: Chemical classes and descriptions from the Northeast Aquatic Habitat Classification System layer. Chemical Class Description 1 Low Buffered, Acidic 2 Moderately Buffered, Neutral 3 Highly Buffered, Calcareous 0 Assume Moderately Buffered (Size 3+ Rivers)

2 Though the initial intent was to include Stream Order as a variable, it was found that the Stream Order values from the Northeast Aquatic Habitat Classification layer were not consistent with the newer National Hydrography stream layer dataset (or the Stream Orders described during the Designated River delineation). It was determined that Watershed Size class was likely to be more influential on flow and fish communities than the Stream Order value over the spatial scale of the dataset.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 13 July, 2018

Table - 9: Temperature classes and descriptions from the Northeast Aquatic Habitat Classification System layer. 3 Temperature Class Description 1 Cold 2 Transitional Cool 3 Transitional Warm 4 Warm

Electrofishing data from Maine, New Hampshire, Vermont, , Rhode Island, Connecticut, and New York were consolidated into a geodatabase. These data included only count data, as CPUE was not readily available from all states. A list of all species in the entire dataset is in Appendix A. Not all species are included in the reference river selection datasets.

From the segments of reference rivers identified, fish community samples were preliminarily selected in GIS. The initial sample selection included locations within a 200-meter buffer of the river segment, and were within areas classified as having a low to very low cumulative disturbance index based on the National Fish Habitat Disturbance Index layer.

If large numbers of samples were initially selected in GIS, then samples were retained based on a HUC4 watershed-level screening. This resulted in the removal of reference river data that were from distant watersheds. Data from more distant reference rivers were used if required to retain a sufficient sample size of fish community data from reference rivers.

The fish sample data were then screened to include only samples:

 Collected from 1990 to the present  Containing 50 individuals or greater  Collected using appropriate survey methods (i.e. General Biological Surveys vs. species-specific sampling), when this information was available  Where more than two samples were present on a reference river

Further screening was performed later as described in the Statistical Analysis section.

3 Note: The temperature classifications in the NAHCS dataset are from a different dataset than the NH fish community types, and the overall values may differ due to different calculation methods. However, the NAHCS data were used for reference river selection because their spatial extent covers the surrounding states.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 14 July, 2018

STATISTICAL ANALYSIS Fish community data from each reference river were evaluated for suitability using Multivariate Pseudo Standard Error (MultSE), as described by Anderson and Santana-Garcon (2015). This method was developed to measure precision for multivariate assemblage data. The concept of MultSE is similar to univariate standard error, but incorporates multivariate community data, with permutation-based means and bias-adjusted bootstrap-based error bars (Anderson and Santana-Garcon 2015). Though the method was developed to evaluate sufficiency of datasets being utilized subsequently for dissimilarity-based null hypothesis testing, it is also useful for determining whether the addition of samples to the dataset would result in considerable changes to the ecological community as a whole.

In this case, we evaluated whether enough samples were present in the fish community dataset to characterize each reference river fish community, and then went further and determined whether enough reference rivers were selected for developing a TFC model. An ideal number of sample sites or reference rivers would yield either a low MultSE value, or provide an asymptotic relationship, with additional sites or reference rivers providing relatively little decrease in MultSE and narrow confidence intervals (see theoretical example in Figure 5). The calculations in the MultSE analyses were based on Bray-Curtis dissimilarity, incorporating the proportional abundance of species, which is directly pertinent to developing the TFC model using the Bain and Meixler (2005) method.

The adequacy of reference river fish community samples was evaluated by calculating MultSE of samples for each reference river (the number of samples per reference river = n). If the MultSE patterns indicated that more samples would have been necessary to characterize the fish community, those reference rivers were removed from further analyses.

After the MultSE screening, the counts from each reference river were summed, consistent with reported methods for the initial calculations in the TFC development used by Bain and Meixler (2005). Another MultSE analysis was performed in the same manner, but using tallied counts for each of the reference rivers (the number of reference rivers = n). If the MultSE analysis yielded an asymptotic function with low MultSE values and narrow error bars, it was determined that the use of these reference river data were sufficient for calculation of a TFC model.

The fish community data from the final selection of reference rivers was then provided in tables, ordered by the rank of mean proportional abundance, which is one of the first calculations for the TFC model development, per Bain and Meixler (2005). It should be noted that a number of species in the reference river data may not be expected to be present in the Designated River segment, due to potential data gathering from distant watersheds; these species are typically low in abundance, and will not be used in the TFC model development phase.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 15 July, 2018

Sufficient Data (use) More Data Beneficial (use if needed) 1 1

0.8 0.8

0.6 0.6 MultSE 0.4 MultSE 0.4

0.2 0.2

0 0 0 2 4 6 8 0 2 4 6 8 n n

Sufficient Data (use) More Data Required (do not use) 1 1

0.8 0.8

0.6 0.6 MultSE MultSE 0.4 0.4

0.2 0.2

0 0 0 2 4 6 8 0 2 4 6 8 n n

Figure - 5: Theoretical example of evaluating the number of samples (i.e. whether data are sufficient for further analysis) using MultSE.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 16 July, 2018

REFERENCE RIVER SELECTION RESULTS Results are provided below for each delineated segment of the Designated River. UPPER SEGMENT Reference rivers were selected for the upper segment of the Pemigewasset River using the characteristics shown in Table - 10.

Table - 10: Characteristics used to select reference rivers for the upper segment. Characteristic Class Description Size Class 1b Creek Elevation Class 3 800 - 1,700 feet Gradient Class 4-5 Moderate High to High Chemical Class 1 Low Buffered (Acidic) Temperature Class 1 Cold Level III Ecoregion 58 Northeastern Highlands

Based on the amount of data selected initially, data from three watersheds, the Watershed (HUC4 0104), the Merrimack River Watershed (HUC4 0107), and the Saco/Coastal Watershed (HUC4 0106) were considered. From this selection and the available fish community data, six reference rivers were identified (Figure – 6). The initial reference rivers identified were:

 Mad River (5 samples)  (4 samples)  South Branch Baker River (2 samples)  (8 samples)  Unnamed Brook - North Wilmot (2 samples)  Whiteface River (2 samples)

Based on the MultSE analysis (Figure - 7), six reference rivers provided either low MultSE values with relatively few site locations, or an asymptotic relationship when more sites were available. No reference rivers were removed from further analysis.

The remaining six reference rivers were then analyzed together using MultSE (Figure - 8). MultSE values, with decreasing confidence intervals, and an asymptotic relationship with increasing numbers of reference rivers, indicated that these reference rivers would be suitable for development of a TFC model. The combined count data for each species and reference river is shown in Table – 11.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 17 July, 2018

Figure - 6: Initial selection of reference river fish community samples for the upper segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 18 July, 2018

Figure - 7: MultSE (beginning at n=2) of fish community data for reference rivers initially selected for the upper segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 19 July, 2018

Figure - 8: MultSE for the final reference river selection for the upper segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 20 July, 2018

Table - 11: Species counts for reference rivers for the upper segment.

Unnamed South Swift Brook - Rank of Nubanusit Branch Diamond North Whiteface Mean Mean Species Mad River Brook Baker River River Wilmot River Proportion Proportion Blacknose Dace 30 42 140 185 225 93 0.356 1 Brook Trout 270 5 4 812 103 82 0.324 2 Slimy Sculpin 60 0 0 0 0 82 0.073 3 Longnose Dace 11 80 18 25 0 0 0.052 4 Fallfish 0 143 0 0 0 0 0.050 5 Longnose Sucker 28 0 17 77 0 2 0.038 6 Atlantic Salmon 92 0 0 0 0 0 0.031 7 Common Shiner 0 75 1 0 0 0 0.027 8 White Sucker 0 55 1 12 0 0 0.022 9 Yellow Bullhead 0 49 0 0 0 0 0.017 10 Tessellated Darter 0 19 0 0 0 0 0.007 11 Rainbow Trout 1 4 0 0 0 0 0.002 12 Smallmouth Bass 0 3 0 0 0 0 0.001 13 Chain Pickerel 0 2 0 0 0 0 0.001 14.5 Yellow Perch 0 2 0 0 0 0 0.001 14.5 Golden Shiner 0 1 0 0 0 0 0.000 16.5 Largemouth Bass 0 1 0 0 0 0 0.000 16.5

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 21 July, 2018

MIDDLE SEGMENT Reference rivers were selected for the middle segment of the Pemigewasset River using the characteristics shown in Table - 12.

Table - 12: Characteristics used to select reference rivers for the middle segment. Characteristic Class Description Size Class 2-3a Small River to Medium Tributary River Elevation Class 2 20 - 800 feet Gradient Class 3 Low Moderate Chemical Class 1-0 Low Buffered (Acidic) to Assume Mod. Buff. (Size 3+) Temperature Class 1-2 Cold to Transitional Cool Level III Ecoregion 58 Northeastern Highlands

Based on the limited amount of data selected initially, data from watersheds east of the Hudson River watershed were considered. From this selection and the available fish community data, nine reference rivers were identified (Figure – 9). The initial reference rivers identified were:

 Baker River (3 samples)  Blackwater River (2 samples)  Crooked River (6 samples)  Deerfield River (7 samples)  Housatonic River (30 samples)  Little Androscoggin River (2 samples)  (4 samples)  South Branch (6 samples)  Williams River (2 samples)

Based on the MultSE analysis (Figure - 10), eight reference rivers provided either low MultSE values with relatively few site locations, or an asymptotic relationship when more sites were available. One reference river (Baker River) was removed from further analysis because the high MultSE value using all samples for these rivers indicated that data may not be sufficient for accurate calculation of proportional abundance in that river. The Little Androscoggin River was removed due to nearly complete dominance of the fish community by Brown Trout in the samples, which was uncharacteristic of any of the other reference rivers.

The remaining eight reference rivers were then analyzed together using MultSE (Figure - 11). MultSE values, with decreasing confidence intervals, and an asymptotic relationship with increasing numbers of reference rivers, indicated that these reference rivers would be suitable for development of a TFC model. The combined count data for each species and reference river is shown in Table – 13.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 22 July, 2018

Figure - 9: Initial selection of reference river fish community samples for the middle segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 23 July, 2018

Figure - 10: MultSE (beginning at n=2) of fish community data for reference rivers initially selected for the middle segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 24 July, 2018

Figure - 11: MultSE for the final reference river selection for the middle segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 25 July, 2018

Table - 13: Species counts for reference rivers for the middle segment.

South Branch Rank of Blackwater Crooked Deerfield Housatonic Millers Piscataquog Williams Mean Mean Species River River River River River River River Proportion Proportion Blacknose Dace 41 2 749 10 3 425 1990 0.231 1 Longnose Dace 27 0 499 1656 20 293 460 0.134 2 Atlantic Salmon 0 519 2 0 0 26 13 0.125 3 Smallmouth Bass 0 20 34 5098 111 25 0 0.097 4 White Sucker 45 5 565 1015 52 15 5 0.092 5 Fallfish 20 22 4 1095 213 33 0 0.084 6 Common Shiner 22 11 340 125 125 73 154 0.081 7 Slimy Sculpin 1 0 714 0 0 0 6 0.033 8 Redbreast Sunfish 1 0 0 85 111 0 0 0.024 9 Margined Madtom 0 0 0 0 0 114 0 0.015 10 Creek Chub 0 2 97 2 0 0 94 0.010 11 Sea Lamprey 0 0 1 0 47 0 0 0.009 12 Tessellated Darter 0 0 98 336 3 0 0 0.009 13 American Eel 0 20 16 0 16 0 0 0.009 14 Rock Bass 0 0 8 526 0 0 0 0.007 15 Yellow Bullhead 0 0 1 1 8 27 0 0.005 16 Brown Trout 0 1 21 270 0 2 0 0.005 17 Bluntnose Minnow 0 0 0 362 0 0 0 0.005 18 Burbot 3 3 0 0 0 0 0 0.003 19 Brook Trout 0 8 6 0 0 9 0 0.003 20 Bluegill 0 0 0 166 5 1 0 0.003 21 Largemouth Bass 1 3 0 43 0 8 0 0.003 22 Rainbow Trout 0 0 0 151 1 3 0 0.003 23 Longnose Sucker 0 0 13 1 0 13 0 0.002 24

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 26 July, 2018

South Branch Rank of Blackwater Crooked Deerfield Housatonic Millers Piscataquog Williams Mean Mean Species River River River River River River River Proportion Proportion Spottail Shiner 0 0 0 160 0 0 0 0.002 25 Yellow Perch 1 0 1 4 0 0 0 0.001 26 Lepomis Sp 0 0 0 0 0 7 0 0.001 27 Pumpkinseed 0 0 0 14 0 4 0 0.001 28 Lake Chub 0 0 0 0 0 0 13 0.001 29 Golden Shiner 0 0 0 8 0 3 0 0.000 30 Common Carp 0 0 0 38 0 0 0 0.000 31 Chain Pickerel 0 0 0 1 1 2 0 0.000 32 Creek Chubsucker 0 0 9 0 0 0 0 0.000 33 Green Sunfish 0 0 0 24 0 0 0 0.000 34 Brown Bullhead 0 0 0 11 0 0 0 0.000 35 Black Crappie 0 0 0 10 0 0 0 0.000 36 Northern Pike 0 0 0 6 0 0 0 0.000 37 Banded Killifish 0 0 0 2 0 0 0 0.000 38 Fathead Minnow 0 0 0 1 0 0 0 0.000 39

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 27 July, 2018

LOWER SEGMENT Reference rivers were selected for the lower segment of the Pemigewasset River using the characteristics shown in Table - 14.

Table - 14: Characteristics used to select reference rivers for the lower segment. Characteristic Class Description Size Class 3a Medium Tributary River Elevation Class 2 20 - 800 feet Gradient Class 1-2 Very Low to Low Chemical Class 0 Assume Mod. Buff (Size 3+) Temperature Class 2 Transitional Cool Level III Ecoregion 58 Northeastern Highlands

Based on the limited amount of data selected initially, no additional selection based on HUC4 watersheds were considered. From this selection and the available fish community data, five reference rivers were identified (Figure – 12). The initial reference rivers identified were:

(4 samples)  Batten Kill (4 samples)  Grass River (2 samples)  Housatonic River (3 samples)  Westfield River (6 samples)

Based on the MultSE analysis (Figure - 13), all five reference rivers provided either low MultSE values with relatively few site locations, or an asymptotic relationship when more sites were available. No reference rivers were removed from further analysis.

The five reference rivers were then analyzed together using MultSE (Figure - 14). Low MultSE values, with narrow confidence intervals, and an asymptotic relationship with increasing numbers of reference rivers, indicated that these reference rivers would be suitable for development of a TFC model. The combined count data for each species and reference river is shown in Table – 15.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 28 July, 2018

Figure - 12: Initial selection of reference river fish community samples for the lower segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 29 July, 2018

Figure - 13: MultSE (beginning at n=2) of fish community data for reference rivers initially selected for the lower segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 30 July, 2018

Figure - 14: MultSE for the final reference river selection for the lower segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 31 July, 2018

Table - 15: Species counts for reference rivers for the lower segment.

Ashuelot Batten Grass Housatonic Westfield Mean Rank of Mean Species River Kill River River River Proportion Proportion Common Shiner 368 7 7 0 506 0.124 1 Bluntnose Minnow 0 1 110 68 0 0.110 2 Fallfish 494 0 0 17 23 0.096 3 Slimy Sculpin 0 420 0 0 0 0.093 4 Blacknose Dace 3 127 45 2 260 0.088 5 White Sucker 55 46 6 90 23 0.073 6 Longnose Dace 112 67 0 3 340 0.072 7 Creek Chub 0 48 77 0 22 0.065 8 Smallmouth Bass 17 0 0 48 318 0.063 9 Rock Bass 3 0 34 44 75 0.055 10 Tessellated Darter 69 29 0 26 87 0.041 11 Pumpkinseed 6 0 0 43 4 0.024 12 American Eel 4 0 0 0 166 0.019 13 Cutlips Minnow 0 76 0 0 0 0.017 14 Yellow Perch 6 0 0 24 1 0.014 15 Brown Trout 1 58 0 0 0 0.013 16 Central Mudminnow 0 0 12 0 0 0.008 17 Yellow Bullhead 23 0 0 0 13 0.005 18 Bluegill 0 0 0 10 0 0.005 19 Redfin Pickerel 0 14 0 0 0 0.003 20 Brook Trout 0 12 0 0 0 0.003 21 Longnose Sucker 0 0 0 3 0 0.002 22 Brown Bullhead 0 0 2 0 0 0.001 23.5 Golden Shiner 0 0 2 0 0 0.001 23.5 Chain Pickerel 2 2 0 0 0 0.001 25 Atlantic Salmon 1 0 0 0 5 0.001 26 Redbreast Sunfish 4 0 0 0 0 0.001 27 Bridle Shiner 0 0 1 0 0 0.001 28 Banded Killifish 0 0 0 1 0 0.001 30.5 Largemouth Bass 0 0 0 1 0 0.001 30.5 Northern Pike 0 0 0 1 0 0.001 30.5 Spottail Shiner 0 0 0 1 0 0.001 30.5 Lake Chub 0 1 0 0 0 0.000 33

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 32 July, 2018

IV. TFC Model Development

TFC MODEL DEVELOPMENT METHODS The TFC model development process included the following steps for each of the Designated River segments:

Develop Fish Species List A comprehensive list of native fish species known to have inhabited the larger-scale basins for each of the Reference Rivers was developed collaboratively between Gomez and Sullivan, NHDES, and NH Fish and Game biologists. These lists were then matched to the Designated River segments that reside within those basins to remove non-native species, and species deemed to be native to the basin were retained for inclusion in the TFC model. Though anadromous species were included initially in the reference river data, and are considered native to most rivers, their abundances would not often be adequately represented by most sampling efforts due to immigration/emigration of individuals. Therefore, anadromous fish (Sea Lamprey; Alewife; Blueback Herring) were removed from TFC model development, consistent with Bain and Meixler (2005). Atlantic Salmon were retained in the analysis because juveniles of this species would typically reside in streams as parr and smolts for at least one year.

Remove Stocked Fish The dataset developed during the reference river data selection phase was evaluated in detail to determine whether stocked fish of native species were present in the catch data. The objective was to remove stocked individuals from the reference river dataset. Removing these individuals was accomplished by using the available metadata and consulting with the state agencies that manage the original fish sample data. Sample-specific information varied among the State datasets; therefore methods for stocked fish removal varied, and included:

 Evaluation of length distributions (NH)  Removal of Brook Trout over 200mm (MA)  Removal of Brook Trout and Atlantic Salmon where no natural reproduction of these species occurs  Removal based on wild/stocked information as available in the dataset

Develop the TFC Models The TFC models were developed from the final dataset using steps adapted from Bain and Meixler (2005), which included:

1. The catch for each species from each sample within a reference river was summed across all samples. 2. Proportions of catch for each species was then summed across Reference Rivers. 3. The summed proportions were ranked by dominance, with a value of “1” being assigned to the most commonly dominant species. Ranks increased with decreasing dominance. 4. The expected proportions of species was calculated by converting the species ranks to reciprocals (1/rank), summing the reciprocal ranks, and then dividing the reciprocal rank by the sum of all of the reciprocal ranks.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 33 July, 2018

The habitat use classification, pollution tolerance, and preferred thermal regime was also shown for each species based on Bain and Meixler (2000) and Yoder et al. (2016).

TFC MODEL RESULTS

SPECIES LIST The comprehensive list of native, resident fish species list for this Designated River, as determined from the Pemigewasset River watershed, is shown in Table 16.

Table - 16: Comprehensive list of native species used for the Designated River watershed, as determined from the greater basin area.

Habitat Use Pollution Species Classification Tolerance Thermal Regime American Eel MG T Eurythermal Atlantic Salmon FS I Cold Banded Killifish MG T Warm Banded Sunfish MG M Warm Blacknose Dace FS T Eurythermal Bridle Shiner MG I Warm Brook Trout FS I Cold Brown Bullhead MG T Warm Burbot FD S Cold Chain Pickerel MG M Warm Common Shiner FD M Eurythermal Creek Chub FS T Eurythermal Creek Chubsucker FS I Eurythermal Fallfish FS M Eurythermal Golden Shiner MG T Eurythermal Lake Chub FD I Cold Longnose Dace FS M Eurythermal Longnose Sucker FD I Cold Northern Redbelly Dace MG I Warm Pumpkinseed MG M Warm Redbreast Sunfish MG M Warm Redfin Pickerel MG M Warm Slimy Sculpin FS I Cold Spottail Shiner MG M Eurythermal Tessellated Darter FS M Eurythermal White Sucker FD T Eurythermal Yellow Perch MG M Eurythermal

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 34 July, 2018

*Note: For Habitat Use Classification – MG = Macrohabitat Generalist; FD = Fluvial Dependent; FS = Fluvial Specialist; E = Estuarine. For Pollution Tolerance – I = Intolerant; S = Sensitive (Moderately Intolerant); M = Moderate Tolerance; T = Tolerant. Information in brackets was not found in Bain and Meixler (2000) or Yoder et al. (2016), and was inserted based on relevant species information.

TFC model results are provided below for each delineated segment of the Designated River.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 35 July, 2018

UPPER SEGMENT The Target Fish Community of the upper delineated segment of the Designated River is shown in Figure 16 and Table 17.

Figure - 16: Graphical representation of the TFC Model for the upper segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 36 July, 2018

Table - 17: Count of fish from reference river data and expected percentage (TFC Model) of species for the upper segment.

South Unnamed Branch Swift Brook Rank of Mad Nubanusit Baker Diamond North Whiteface Mean Mean Expected Species River Brook River River Wilmot River Proportion Proportion Percentage Blacknose Dace 30 42 140 185 225 93 0.37358 1 32.2% Brook Trout 270 0 3 812 70 82 0.32969 2 16.1% Slimy Sculpin 60 0 0 0 0 82 0.07783 3 10.7% Fallfish 0 143 0 0 0 0 0.05688 4 8.1% Longnose Dace 11 80 18 25 0 0 0.05683 5 6.4% Longnose Sucker 28 0 17 77 0 2 0.04027 6 5.4% Common Shiner 0 75 1 0 0 0 0.03076 7 4.6% White Sucker 0 55 1 12 0 0 0.02460 8 4.0% Tessellated Darter 0 19 0 0 0 0 0.00756 9 3.6% Chain Pickerel 0 2 0 0 0 0 0.00080 10.5 3.1% Yellow Perch 0 2 0 0 0 0 0.00080 10.5 3.1% Golden Shiner 0 1 0 0 0 0 0.00040 12 2.7%

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 37 July, 2018

MIDDLE SEGMENT The Target Fish Community of the middle delineated segment of the Designated River is shown in Figure 17 and Table 18.

Figure - 17: Graphical representation of the TFC Model for the middle segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 38 July, 2018

Table - 18: Count of fish from reference river data and expected percentage (TFC Model) of species for the middle segment. South Branch Rank of Blackwater Crooked Deerfield Housatonic Millers Piscataquog Williams Mean Mean Expected Species River River River River River River River Proportion Proportion Percentage Blacknose Dace 41 2 749 10 3 425 1990 0.24732 1 26.8% Longnose Dace 27 0 499 1656 20 293 460 0.17716 2 13.4% Atlantic Salmon 0 519 0 0 0 0 0 0.12524 3 8.9% Fallfish 20 22 4 1095 213 33 0 0.11922 4 6.7% White Sucker 45 5 565 1015 52 15 5 0.11553 5 5.4% Common Shiner 22 11 340 125 125 73 154 0.09475 6 4.5% Slimy Sculpin 1 0 714 0 0 0 6 0.03399 7 3.8% Redbreast Sunfish 1 0 0 85 111 0 0 0.03272 8 3.3% Tessellated Darter 0 0 98 336 3 0 0 0.01590 9 3.0% Creek Chub 0 2 97 2 0 0 94 0.00993 10 2.7% American Eel 0 20 16 0 16 0 0 0.00976 11 2.4% Spottail Shiner 0 0 0 160 0 0 0 0.00505 12 2.2% Burbot 3 3 0 0 0 0 0 0.00339 13 2.1% Longnose Sucker 0 0 13 1 0 13 0 0.00279 14 1.9% Brook Trout 0 8 6 0 0 0 0 0.00221 15 1.8% Pumpkinseed 0 0 0 14 0 4 0 0.00111 16 1.7% Yellow Perch 1 0 1 4 0 0 0 0.00106 17 1.6% Golden Shiner 0 0 0 8 0 3 0 0.00075 18 1.5% Lake Chub 0 0 0 0 0 0 13 0.00068 19 1.4% Chain Pickerel 0 0 0 1 1 2 0 0.00063 20 1.3% Creek Chubsucker 0 0 9 0 0 0 0 0.00041 21 1.3% Brown Bullhead 0 0 0 11 0 0 0 0.00035 22 1.2% Banded Killifish 0 0 0 2 0 0 0 0.00006 23 1.2%

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 39 July, 2018

LOWER SEGMENT The Target Fish Community of the lower delineated segment of the Designated River is shown in Figure 18 and Table 19.

Figure - 18: Graphical representation of the TFC Model for the lower segment.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 40 July, 2018

Table - 19: Count of fish from reference river data and expected percentage (TFC Model) of species for the lower segment.

Rank of Ashuelot Batten Grass Housatonic Westfield Mean Mean Expected Species River Kill River River River Proportion Proportion Percentage Common Shiner 368 7 7 0 506 0.14802 1 27.1% Blacknose Dace 3 127 45 2 260 0.13590 2 13.5% Creek Chub 0 48 77 0 22 0.12549 3 9.0% White Sucker 55 46 6 90 23 0.11919 4 6.8% Slimy Sculpin 0 420 0 0 0 0.10867 5 5.4% Fallfish 494 0 0 17 23 0.10738 6 4.5% Longnose Dace 112 67 0 3 340 0.08762 7 3.9% Tessellated Darter 69 29 0 26 87 0.05670 8 3.4% Pumpkinseed 6 0 0 43 4 0.04258 9 3.0% Yellow Perch 6 0 0 24 1 0.02407 10 2.7% American Eel 4 0 0 0 166 0.02390 11 2.5% Redfin Pickerel 0 14 0 0 0 0.00362 12 2.3% Brook Trout 0 12 0 0 0 0.00310 13 2.1% Brown Bullhead 0 0 2 0 0 0.00286 15 1.8% Golden Shiner 0 0 2 0 0 0.00286 15 1.8% Longnose Sucker 0 0 0 3 0 0.00286 15 1.8% Bridle Shiner 0 0 1 0 0 0.00143 17 1.6% Banded Killifish 0 0 0 1 0 0.00095 18.5 1.5% Spottail Shiner 0 0 0 1 0 0.00095 18.5 1.5% Chain Pickerel 2 2 0 0 0 0.00087 20 1.4% Redbreast Sunfish 4 0 0 0 0 0.00071 21 1.3% Lake Chub 0 1 0 0 0 0.00026 22 1.2%

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 41 July, 2018

V. References Cited

Anderson, M.J. and J. Santana-Garcon. 2015. Measures of precision for dissimilarity-based multivariate analysis of ecological communities. Ecology Letters 18: 66-73.

Bain, M.B. and M.S. Meixler. 2000. Defining a target fish community for planning and evaluating enhancement of the Quinebaug River in Massachusetts and Connecticut. Developed for Quinebaug River Instream Flow Study Agencies, July 2000.

Bain, M.B. and M.S. Meixler. 2005. Defining a target fish community for planning and evaluating river rehabilitation. Manuscript submitted to Environmental Biology of Fishes in April 2005.

Clarke, K.R. 1993. Nonparametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18(1): 117-143.

Hansen, M.J., T.D. Beard, and D.B. Hayes. 2007. Sampling and experimental design. Pages 51-120 in C.S. Guy and M.L. Brown, editors. Analysis and Interpretation of Freshwater Fisheries Data. American Fisheries Society, Bethesda, MD.

Olivero, A.P. and M.G. Anderson. 2008. Northeast Aquatic Habitat Classification System. The Nature Conservancy, Eastern Regional Office. September 30, 2008.

U.S. Environmental Protection Agency (USEPA). 2013. Primary distinguishing characteristics of Level III Ecoregions of the continental United States. September, 2013. ftp://newftp.epa.gov/EPADataCommons/ORD/Ecoregions/us/Eco_Level_III_descriptions.doc

Yoder, C.O., Thoma, R.F., and L.E. Hersha. 2016. Maine Rivers Fish Assemblage Assessment: Development of an Index of Biotic Integrity for Non-Wadeable Rivers. MBI Technical Report MBI/2008-22—2. March 8, 2009, Addendum March 31, 2016.

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 42 July, 2018

Appendix A

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 43 July, 2018

Table – A1: List of common and scientific names for fish species in the fish community sample dataset (includes samples from NY, CT, RI, MA, VT, NH, and ME).

Common Name Scientific Name Alewife Alosa pseudoharengus Allegheny Pearl Dace Margariscus margarita American Brook Lamprey Lampetra appendix American Eel Anguilla rostrata American Shad Alosa sapidissima Atlantic Salmon Salmo salar Banded Darter Etheostoma zonale Banded Killifish Fundulus diaphanus Banded Sunfish Enneacanthus obesus Bigeye Chub Hybopsis amblops Bigmouth Buffalo Ictiobus cyprinellus Bigmouth Shiner Notropis dorsalis Black Bullhead Ameiurus melas Black Crappie Pomoxis nigromaculatus Black Redhorse Moxostoma duquesnei Blackchin Shiner Notropis heterodon Blacknose Dace Rhinichthys atratulus Blacknose Shiner Notropis heterolepis Blackside Darter Percina maculata Blueback Herring Alosa aestivalis Bluebreast Darter Etheostoma camurum Bluegill Lepomis macrochirus Bluespotted Sunfish Enneacanthus gloriosus Bluntnose Minnow Pimephales notatus Bowfin Amia calva Brassy Minnow Hybognathus hankinsoni Bridle Shiner Notropis bifrenatus Brindled Madtom Noturus miurus Brook Silverside Labidesthes sicculus Brook Stickleback Culaea inconstans Brook Trout Salvelinus fontinalis Brown Bullhead Ameiurus nebulosus Brown Trout Salmo trutta Burbot Lota lota Central Mudminnow Umbra limi Central Stoneroller Campostoma anomalum

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 44 July, 2018

Common Name Scientific Name Chain Pickerel Esox niger Channel Catfish Ictalurus punctatus Channel Darter Percina copelandi Chinook Salmon Oncorhynchus tshawytscha Coho Salmon Oncorhynchus kisutch Comely Shiner Notropis amoenus Common Carp Cyprinus carpio Common Shiner Luxilus cornutus Creek Chub Semotilus atromaculatus Creek Chubsucker Erimyzon oblongus Cutlips Minnow Exoglossum maxillingua Eastern Mudminnow Umbra pygmaea Eastern Sand Darter Ammocrypta pellucida Eastern Silvery Minnow Hybognathus regius Emerald Shiner Notropis atherinoides Fallfish Semotilus corporalis Fantail Darter Etheostoma flabellare Fat Sleeper Goby Dormitator maculatus Fathead Minnow Pimephales promelas Finescale Dace Phoxinus neogaeus Fourspine Stickleback Apeltes quadracus Freshwater Drum Aplodinotus grunniens Gizzard Shad Dorosoma cepedianum Golden Redhorse Moxostoma erythrurum Golden Shiner Notemigonus crysoleucas Goldfish Carassius auratus Grass Carp Ctenopharyngodon idella Grass Pickerel Esox americanus vermiculatus Gravel Chub Erimystax x-punctatus Greater Redhorse Moxostoma valenciennesi Green Sunfish Lepomis cyanellus Greenside Darter Etheostoma blennioides Hickory Shad Alosa mediocris Hogchoker Trinectes maculatus Hornyhead Chub Nocomis biguttatus Inland Silverside Menidia beryllina Iowa Darter Etheostoma exile Johnny Darter Etheostoma nigrum

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 45 July, 2018

Common Name Scientific Name Koi Cyprinus rubrofuscus Kokanee/Sockeye Salmon Oncorhynchus nerka Lake Chub Couesius plumbeus Lake Chubsucker Erimyzon sucetta Lake Sturgeon Acipenser fulvescens Lake Trout Salvelinus namaycush Landlocked Salmon Salmo salar Largemouth Bass Micropterus salmoides Logperch Percina caprodes Longhead Darter Percina macrocephala Longnose Dace Rhinichthys cataractae Longnose Gar Lepisosteus osseus Longnose Sucker Catostomus catostomus Margined Madtom Noturus insignis Mimic Shiner Notropis volucellus Mooneye Hiodon tergisus Mottled Sculpin Cottus bairdii Mountain Brook Lamprey Ichthyomyzon greeleyi Mud Sunfish Acantharchus pomotis Mummichog Fundulus heteroclitus Muskellunge Esox masquinongy Ninespine Stickleback Pungitius pungitius Northern Brook Lamprey Ichthyomyzon fossor Northern Hog Sucker Hypentelium nigricans Northern Pike Esox lucius Northern Redbelly Dace Phoxinus eos Northern Snakehead Channa argus Northern Sunfish Lepomis megalotis Ohio Lamprey Ichthyomyzon bdellium Oriental Weatherfish Misgurnus anguillicaudatus Pearl Dace Margariscus sp Pirate Perch Aphredoderus sayanus Pugnose Shiner Notropis anogenus Pumpkinseed Lepomis gibbosus Quillback Carpiodes cyprinus Rainbow Darter Etheostoma caeruleum Rainbow Smelt Osmerus mordax Rainbow Trout Oncorhynchus mykiss

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 46 July, 2018

Common Name Scientific Name Redbreast Sunfish Lepomis auritus Redear Sunfish Lepomis microlophus Redfin Pickerel Esox americanus americanus Redfin Shiner Lythrurus umbratilis Redside Dace Clinostomus elongatus River Chub Nocomis micropogon River Redhorse Moxostoma carinatum Rock Bass Ambloplites rupestris Rosyface Shiner Notropis rubellus Rosyside Dace Clinostomus funduloides Round Goby Neogobius melanostomus Round Whitefish Prosopium cylindraceum Rudd Scardinius erythrophthalmus Sand Shiner Notropis stramineus Satinfin Shiner Cyprinella analostana Sauger Sander canadensis Sea Lamprey Petromyzon marinus Sheepshead Minnow Cyprinodon variegatus Shield Darter Percina peltata Shorthead Redhorse Moxostoma macrolepidotum Shortnose Sturgeon Acipenser brevirostrum Silver Lamprey Ichthyomyzon unicuspis Silver Redhorse Moxostoma anisurum Silver Shiner Notropis photogenis Silverjaw Minnow Notropis buccatus Slimy Sculpin Cottus cognatus Smallmouth Bass Micropterus dolomieu Smallmouth Redhorse Moxostoma breviceps Salvelinus fontinalis x Splake namaycush Spotfin Shiner Cyprinella spiloptera Spottail Shiner Notropis hudsonius Spotted Darter Etheostoma maculatum Spotted Sucker Minytrema melanops Stonecat Noturus flavus Streamline Chub Erimystax dissimilis Striped Bass Morone saxatilis Striped Killifish Fundulus majalis Striped Mullet Mugil cephalus

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 47 July, 2018

Common Name Scientific Name Striped Shiner Luxilus chrysocephalus Summer Sucker Catostomus Utawana Swallowtail Shiner Notropis procne Swamp Darter Etheostoma fusiforme Tadpole Madtom Noturus gyrinus Tench Tinca tinca Tessellated Darter Etheostoma olmstedi Threespine Stickleback Gasterosteus aculeatus Tidewater Silverside Menidia beryllina Tiger Muskellunge Esox lucius x masquinongy Salmo trutta x Salvelinus Tiger Trout fontinalis Tonguetied Minnow Exoglossum laurae Trout Perch Percopsis omiscomaycus Variegate Darter Etheostoma variatum Walleye Sander vitreus Warmouth Lepomis gulosus Western Blacknose Dace Rhinichthys obtusus Western Mosquitofish Gambusia affinis White Bass Morone chrysops White Catfish Ameiurus catus White Crappie Pomoxis annularis White Mullet Mugil curema White Perch Morone americana White Sucker Catostomus commersonii Wreckfish Polyprion americanus Yellow Bullhead Ameiurus natalis Yellow Perch Perca flavescens

NH Target Fish Community Assessment Gomez and Sullivan Engineers Final Pemigewasset River Report 48 July, 2018